The overall goals of this project are to understand how polyoma virus RNAs are made and regulated in infected cells and to use this virus as a model system to learn more general rules for mammalian RNA processing and function. We have discovered that much regulation of viral gene expression depends on the formation of double-stranded RNAs and their editing by the cellular dsRNA specific adenosine deaminase (ADAR1). Experiments completed during the previous funding period suggested that the long-sought trigger for the viral early-late switch might be the overlap and perhaps also the editing of the early and late polyadenylation signals. This represents a new mechanism of gene regulation. We also learned much new about nuclear responses to dsRNAs, including the involvement of a long nuclear-retained noncoding RNA (NEAT1) in the assembly of nuclear bodies called paraspeckles, which function in the retention of edited RNAs. We will extend these studies in three related aims. In the first, we will use new methods to better characterize RNA expression in the viral life cycle and will determine the nature and interplay of DNA sequence elements that promote regulated gene expression. In the second aim we will examine more closely the fate of polyoma dsRNAs in the nucleus, and the impact of cellular dsRNA response pathways including paraspeckles on infection. In the final aim we will carry out studies to learn more about how antisense RNA works in mammalian nuclei. This work will involve studying whether single DNA molecules can allow concurrent transcription in both directions as well as a new approach using small molecule-directed heterodimerization to alter the intranuclear proximity of DNA and RNA molecules.